We propose to develop a theoretically and ethnographically informed model of HIV risk environment for public health intervention by identifying the key structural forces, institutional processes, and socio-cultural constraints that shape HIV behaviors among inner-city street-based drug users. We will examine the social processes that drive the logics for HIV transmission behaviors among street-based injection drug users (IDUs) in two cities, Philadelphia and San Francisco, that span a strategic range of the demographic, economic, political, and cultural factors affecting inner-city poverty in the United States. Building on long-term, continuous participant-observation ethnographic fieldwork in the shooting encampments of ethnically and generationally bounded social networks of heroin injectors and crack smokers in San Francisco as well as pilot research in Philadelphia at open-air drug markets and shooting galleries, we will document patterns of: 1) law enforcement and criminal justice; 2) poverty and income generating opportunities (including legal and illegal labor markets and public subsidies); 3) illegal drug markets and polydrug availability; 4) racialized ethnic boundaries and conflicts; 5) gender/sexuality boundaries that delimit social networks; 6) oppositional sub-cultural identities that trap IDUs into counterproductive relationships with the services that are meant to help or contain them; 7) violence and insecurity; 8) access to clinical care, public health prevention messages, and social services; 9) access to ecological and basic survival needs infrastructures including transportation systems, food, and shelter from inclement weather. An experienced ethnographic team in both cities will collect data on individual-level HIV risk behaviors at fieldsites that we have successfully maintained on the parent grant for more than 12 years in San Francisco and recently established in Philadelphia (core N of each network=15-20 at any given time, peripheral N=10-15 in each fieldsite; 2 sites in Philadephia and 2 in San Francisco). The project will administer a set of procedures for cross-methodological collaboration with epidemiologists who are the Principal Investigators on long-term, large quantitative data sets documenting street-based drug users in both cities. This will enable us to examine the representativity of our ethnographic data on individual street-based drug users and to explore the generalizability of our analysis of the larger social structural processes that promote or constrain risky injection and sexual practices. It will also bring one of the largest and oldest NIDA-funded data sets on IDUs in San Francisco, an exceptionally high-tech, wealthy, culturally diverse, and liberal city into an illustrative contrast and dialogue with another large NIDA-supported quantitative data set of street-based drug users in the city of Philadelphia, one of the nation's poorest deindustrialized metropolises, with a declining population of largely native-born white, Latino (primarily Puerto Rican) and African American ethnic groups.